Ocular inflammatory diseases, including uveitis, scleritis, and mucous membrane pemphigoid, are major blinding eye diseases in the United States. For some patients, corticosteroid therapy is insufficient to control [unreadable] ocular inflammatory disease, requiring immunosuppression with antimetabolite, T-cell inhibitor, and/or alkylating agent therapies. It has been suggested, based on studies of patients with severe immunologic or other systemic diseases, that such treatments may result in an increased risk of cancer and other morbidities. In these studies, it has been difficult to determine whether the excess risk arose from the underlying diseases or the treatment. We propose to evaluate directly whether immunosuppressive therapy for ocular inflammatory diseases is associated with an excess risk of mortality, cancer, and other major diseases. The study is expected to generate critical information in deciding whether immunosuppressive therapy is warranted for such patients, and whether certain such agents should be avoided. [unreadable] [unreadable] The study will have a classic retrospective cohort design. Patients "exposed" to immunosuppressive therapies will be compared to an external standard, the general United Slates population, and to an internal comparison group, patients with the same ocular inflammatory diseases who did not receive immunosuppression. An estimated 4,695 patients will be accrued from three centers which pioneered the use of immunosuppressive therapy for eye diseases, beginning 17-27 years ago. Patients who received immunosuppressive therapy for eye diseases, and patients with the same ocular inflammatory diagnoses who did not, will be identified by chart reviews. Patients who have died will be identified through search of the National Death Index and other record systems. Death certificates will be the basis for cause-specific mortality analyses. Patients still living will be contacted and queried regarding the occurrence of cancer, osteoporotic, and heart disease diagnoses. Next of kin will also be queried for subjects who have died to ascertain whether such diagnoses occurred prior to death. Such diagnoses will be verified by medical record review by an outcomes committee. The primary outcomes-mortality, cause-specific mortality, and cancer/other disease incidence-will be analyzed using a relative incidence approach. Analysis of the beneficial effects of immunosuppression, based on clinical information only, will be a secondary objective. [unreadable] [unreadable]